1. Field of the Invention
The present invention includes a process for the transformation of a 3-enol ether Δ3,5-steroid to the corresponding Δ4,6-3-ketal steroid (I-P).
The present invention includes a process for the transformation of a Δ4,6-3-keto steroid or ketal thereof (I), to the corresponding Δ4-3-ketosteroid-7α-carboxylic acid (VI).
The present invention also includes a novel processes and novel intermediates to produce the pharmaceutically useful compound eplerenone.
Further, the invention includes processes for transformation of 11α-hydroxy-17-lactone (CI) or 11α-hydroxy steroids (CIV) to the corresponding Δ9(11)-17-lactone (CII) or Δ9(11)-steroids (CV) using a N-fluoroalkylamine reagents (CVI).
2. Description of the Related Art
It is known to transform 3-keto-Δ4,6-steroids into the corresponding steroidal Δ4,6-3-ketals by acid-catalyzed ketalization. Yields are moderate and double bond deconjugation can be competitive. For example, Δ4,6-cholestadiene-3-one-3-cycloethyleneketal was prepared by ketalization of Δ4,6-cholestadien-3-one in 64% yield, see J. Org. Chem. 26, 2549 (1961). Also, 17β-hydroxyandrosta-4,6-dien-3-one-3-cycloethyleneketal was prepared by ketalization of 6-dehydrotestosterone in 55% crude yield, see J. Am. Chem. Soc., 86, 2183 (1964). The steroidal Δ4,6-3-ketals (I-P) can be used as starting materials in the process to prepare eplerenone.
J. Org. Chem., 29, 601 (1964) reports that Δ3,5-3-alkoxy steroids react with DDQ in the presence of water to give the corresponding Δ4,6-3-keto steroids. The process of the present invention reacts Δ3,5-3-alkoxy steroids (3-alkyl enol ether) with DDQ in the presence of an alcohol under essentially anhydrous conditions to give the Δ4,6-3-ketal steroid (I-P). In addition, the prior art methods of producing the Δ4,6-3-ketal steroid (I-P) uses two steps, 6-dehydrogenation of an enol ether to a Δ4,6-3-keto steroid followed by ketalization whereas the present invention is a one step reaction.
Eplerenone, also known as epoxymexrenone, is a useful pharmaceutical agent and chemically is 9α,11α-epoxy-17β-hydroxypregn-4-en-3-one-7α,21-dicarboxylic acid, γ-lactone, methyl ester.
International Publication WO98/25948 of PCT application PCT/US97/23090 discloses eplerenone and many different process to prepare eplerenone. In particular, see schemes 1 thru 10.
U.S. Pat. No. 4,874,754 discloses 19-nor steroids with 7α-aryl substitution. The 7α-aryl substituent included a number of groups including phenyl, thienyl, furyl, thiazolyl, pyrrolyl, oxazolyl, imidazolyl, pyrazolyl, triazolyl, tetrazolyl, isothiazolyl and isoxazolyl, pyridinyl, pyridazinyl, pyrimidinyl and pyrazinyl. Regardless of which group was used, the 19-nor compounds had antiproliferative, anti-estrogenic and/or estrogenic properties and are not useful intermediates to eplerenone because there are no practical methods for installing the 19-methyl group into 19-nor steroids. The 7α-substituted steroids (II) of the present invention, are intermediates, not end products and do not have estrogenic properties because they are not 19-nor steroids.
U.S. Pat. No. 4,502,989 discloses a number of Δ11-steroidal-γ-lactones many of which are substituted in the 7α-position which have aldosterone antagonist activity. The 7α-substitution is 6α,7α-methylene-, 7α-trimethylacetylthio-, 7α-acetylthio- and 7α-benzoythio-, see claim 1. These compounds differ from the compounds of the invention in that the C-ring double bond is Δ11- and the 7α-substituents are such that the compounds cannot be used in the same way as the 7α-substituted steroids (II).
Het., 25, 399 (1987) and Bull. Soc. Chim. Fr. 131, 900 (1994) disclose the use of boron trifluoride diethyl etherate to catalyze conjugate addition of non-steroidal 2-methylfuran to α,β-unsaturated ketones in ethanol/nitromethane. The process of the present invention involves steroidal furans. In addition, the enone substrates in Het., 25, 399 (1987) and Bull Soc. Chim. Fr. 131, 900 (1994) do not contain stereocenters, so the issue of stereocontrol does not arise.
Methods for conjugate addition of carbon nucleophiles to 9(11)-saturated-Δ4,6-3-keto steroids to give 9(11)-saturated-7α-substituted steroids stereoselectively are known. J. Am. Chem. Soc., 94, 4654 (1972) discloses conjugate addition of carbon nucleophiles to 9(11)-saturated-Δ4,6-3-keto steroids to give 9(11 )-saturated-7α-substituted steroids stereoselectively. Tet., 49, 9955 (1993) and Tet. Lett., 29, 1533 (1988) disclose stereoselective addition of allyltrimethylsilane to canrenone (titanium tetrachloride, methylene chloride, −78°) to give a mixture of two difficult-to-separate products (7α-allyl-canrenone and the corresponding 6α,7α-fused silylcyclopentane) in poor yields (43–73% and 7–15%, respectively). Note that in these cases the steroid substrate is 9(11) saturated. All attempts to apply these methods or similar methods to 9(11) unsaturated steroid substrates have failed, due to lack of stereocontrol. For example, U.S. Pat. No. 4,559,332, Example 7, discloses that trimethylsulfoxonium iodide adds to Δ9(11)-canrenone (I) using sodium hydride in DMSO at room temperature to give exclusively 6β,7β-methylene-Δ9(11)-canrenone. Also, nitromethane adds to Δ9(11)-canrenone (I) in tetramethylguanidine at room temperature over 7.5 hrs.) to give exclusively the 7β stereoisomer (7β-nitromethyl-Δ9(11)-6,7-dihydrocanrenone.
Helv. Chim. Acta, 80, 566 (1997) and U.S. Pat. No. 4,559,332 disclose that reaction of Δ9(11)-canrenone with diethylaluminum cyanide to give 7α-cyano-Δ9(11)-6,7-dihydrocanrenone, but the crude product is described as a “brownish amorphous residue” that “was filtered through silica gel yielding amorphous” semipurified product “which was used without further purification in the next step.” The ratio of 7-α to 7-β epimers is not disclosed.
J. Am. Chem. Soc. 79, 3120 (1957), J. Am. Chem. Soc. 82, 6136 (1960), and J. Org. Chem. 27, 1192 (1962) disclose degradation of non-steroidal enediones to carboxylic acids through alkoxyhydroperoxide intermediates and not hydroxyhydroperoxide intermediates. The process of the present invention involves steroidal enediones.
The oxidative opening of furans to carboxylic acids, or carboxylic acid derivatives, by direct ozonolysis is known. However, the yields are usually quite poor. J. Org. Chem., 61, 9126 (1996), reported that a 2,5-disubstituted furan on ozonization underwent partial cleavage to an enol acetate rather than complete cleavage to the carboxylic acid. Het., 34, 895 (1992) reported direct ozonization of a 2-substituted furan gave, after esterification, the methyl ester in 59% yield. J. Am. Chem. Soc. 101, 259 (1979) reported direct ozonization of a 2-substituted furan gave, after esterification, the methyl ester in 55% yield. J. Am. Chem. Soc., 107, 7762 (1985) reported direct ozonization of a 2-sugar-substituted furan gave, after borane reduction, the primary alcohol in 50% yield. Tet. Lett., 34, 7323 (1993) reported direct ozonization of a 2-substituted furan gave, after esterification, the methyl ester in 60% yield. Carb. Res., 150, 163 (1986) reported direct ozonization of a 2-sugar-substituted furan afforded, after reduction with triphenylphosphine followed by lithium aluminum hydride, the primary alcohol in 11% yield. Tet. Lett., 22, 141 (1981) reported direct ozonization of a 2-substituted furan gave, after oxidative workup, the carboxylic acid in approximately 30% yield. J. Am. Chem. Soc., 109, 2082 (1987) reported direct ozonization of a 2-substituted furan gave, after esterification, the methyl ester in 77% yield. Tet. Lett., 39, 7013 (1998) reported direct ozonization of a 2-substituted furan gave, after esterification, the methyl ester in 78%–87% yield. J. Org. Chem., 54, 2085 (1989) reported direct ozonization of two 2-substituted furans gives the carboxylic acid in 89–95% yield, however, in this study, the 2-substituted furans were very simple (i.e., they did not contain any reactive functional group other than the furan). There is no disclosures of a two step furan opening and then oxidative cleavage to the carboxylic acid which results in high yields.
J. Org. Chem. 63, 7505 (1998) discloses the use of dibromatin, sodium bicarbonate and aqueous acetone to open non-steroidal furans to produce enediones. The process of the present invention involves steroidal furans.
Chem. Lett., 1771 (1983) discloses the use of hydrochloric acid in ether to catalyze the isomerization of non-steroidal cis-enediones to trans-enediones. The process of the present invention involves steroidal enediones.
J. Am. Chem. Soc., 79, 3120 (1957), J. Am. Chem. Soc., 82, 6136 (1960) and J. Org. Chem., 27, 1192 (1962) disclose the degradation of enediones to carboxylic acids through alkoxyhydroperoxide intermediates by use of ozone and an oxidatively cleaving agent. The yields are not particularly high. For example, the yield of benzoic acid from trans-dibenzoylethylene was 54%. Following this process, methoxyhydroperoxide (IV-OOH) (where R7-2═—CH3) gave a 65.2/34.8 mixture of the desired carboxylic acid (VI) and α-ketomethylester where (Rb═OMe). The α-ketomethyl ester can not be transformed to an eplerenone useful compound and its production makes this process not commercially useful. By contrast, in the process of this invention, the enedione (III) is degraded to the carboxylic acid (VI) through the hydroxyhydroperoxide intermediate (IV-OOH, where R7-2═—H), which surprisingly rearranges to the desired carboxylic acid (VI) in nearly quantitative yield. The process of the present invention uses ozone, a hydroperoxy-deoxygenating agent and then a oxidatively cleaving agent to avoid production of the α-ketomethylester and obtain increased yields.
Drugs of the Future, 24, 488 (1999) discloses conversion of the 5,7-lactone (VII) to the corresponding methyl ester (VIII) by treatment with “methyl iodide in basic medium,”. The process of the present invention for methylation is a sequential process.
International Publication WO98/25948 generically discloses (5,7)-17-bislactones and 3 protected forms.
International Publication WO98/25948 discloses the transformation of a steroidal 7α-acid to the (5,7)-17-bislactone. This process requires an orthoester. The process of the present invention does not require an orthoester.
International Publication WO98/25948 discloses the transformation of a (5,7)-17-bislactone to the corresponding 7α—CO—OCH3 in one step. The present invention uses two steps but obtains better yields and consumes less reagent.
Eplerenone is 9(11 )α-epoxy-17β-hydroxypregn-4-en-3-one-7α,21-dicarboxylic acid, γ-lactone, methyl ester and as such contains a 7α-carbomethoxy substituent. From the standpoint of production, a major difficulty in the production of eplerenone is introduction of the 7α-carbomethoxy substituent. The present invention includes an improved proved process for the introduction of the 7α-substituent.
It is known that a carboxylic acid can be obtained from a (substituted)furan in one step by ozonolysis. However, the yields are quite low. Further, it is known that furans can be opened to enediones. It is also known that enediones can be oxidized to carboxylic acids.
Bulletin of the Chemical Society of Japan, 52, 3377–3380 (1979) discloses that N-(1,1,2,2,3,3,3)hexafluoropropyldiethylamine, “Ishikawa reagent” is used to replace a hydroxyl group with a fluorine atom or eliminate a hydroxyl group to an olefin. With cyclohexanol, a simple monocyclic system, the elimination product olefin was 78%. However, when the “Ishikawa reagent” was applied to a steroid, cholesterol, the corresponding fluoro compound cholesteryl fluoride was obtained in 83% yield; no elimination product was reported.
J. Org. Chem., 2187–2195(1964) discloses the reaction of 11α-hydroxypregn-4-ene-3,20-dione with 2-chloro-1,1,2-trifluorotriethylamine to give the elimination product, pregna-4,9(11)-diene-3,20-dione, in 86% yield. The process of the present invention does not use 2-chloro-1,1,2-trifluorotriethylamine also known as Yarovenko reagent. Further, use of 2-chloro-1,1,2-trifluorotriethylamine is a problem because it is not stable enough to make scale up practicable. In addition, it is derived from a chlorofluorocarbon and is not environmentally sound.
Tetrahedron Letters, 1065–1069 (1962) also discloses the reaction of 11α-hydroxypregn-4-ene-3,20-dione with 2-chloro-1,1,2-trifluorotriethylamine to give the elimination product, pregna-4,9(11)-diene-3,20-dione.
J. Fluorine Chem., 109, 25–31 (2001) describes and compares the use of 1,1,2,2-tetrafluoroethyl-N,N-dimethylamine as well as Yarovenko-Raksha and Ishikawa reagent as fluorinating and dehydrating agents. While the document discloses examples of elimination reactions in both aliphatic and cyclic systems, the primary use is as a fluorinating agent. The only steroid example was the reaction of 1,1,2,2-tetrafluoroethyl-N,N-dimethylamine with cholesterol which produced a product with fluorine at the C-3 position of cholesterol.